the e and the cpx signal transduction systems control the...

The E and the Cpx signal transduction systems control the synthesis of periplasmic protein-folding enzymes in Escherichia coli

Paul N. D a n e s e I and T h o m a s J. S i lhavy 2

Department of Molecular Biology, Princeton University, Princeton, New Jersey 08544 USA

In Escherichia coli, the heat shock-inducible ,r-factor ~E and the Cpx two-component signal transduction system are both attuned to extracytoplasmic stimuli. For example, cr F activity rises in response to the overproduction of various outer-membrane proteins. Similarly, the activity of the Cpx signal transduction pathway, which consists of an inner-membrane sensor (CpxA) and a cognate response regulator (CpxR), is stimulated by overproduction of the outer-membrane lipoprotein, NIpE. In response to these extracytoplasmic stimuli, ~r E and CpxA/CpxR stimulate the transcription of degP, which encodes a periplasmic protease. This suggests that CpxA/CpxR and tr E both mediate protein turnover within the bacterial envelope. Here, we show that CpxA/CpxR and cr E also control the synthesis of periplasmic enzymes that can facilitate protein-folding reactions. Specifically, cr E controls transcription of fkpA, which specifies a periplasmic peptidyl-prolyl cis/trans isomerase. Similarly, the Cpx system controls transcription of the dsbA locus, which encodes a periplasmic enzyme required for efficient disulfide bond formation in several extracytoplasmic proteins. Taken together, these results indicate that ,rE and CpxA/CpxR are involved in regulating both protein-turnover and protein-folding activities within the bacterial envelope.

[Key Words: Molecular chaperone; stress response; receptor kinase]

Received January 24, 1997; revised version accepted March 27, 1997.

The activity of ¢E, the second heat shock-inducible c-factor in Escherichia coli, is attuned to the physiology of E. coli's extracytoplasmic compartments. For in- stance, Mecsas et al. (1993) have shown that cE activity is modulated in response to the level of outer-membrane proteins found in the outer membrane. Additionally, strains lacking cE (rpoE-) are hypersensitive to deter- gents and hydrophobic agents, suggesting an outer-membrane-permeability defect (Raina et al. 1995; Rouvihre et al. 1995). Furthermore, cE is most homologous to a family of c-factors that is involved in regulating extracytoplasmic and extracellular functions (Lonetto et al. 1994; Raina et al. 1995; Rouvi6re et al. 1995). Taken together, these results intimate a connection between cE and E. coli's extracytoplasmic regions.

Consistent with this view, one of the major regulatory targets of cE is degP, which encodes a periplasmic protease that degrades various aberrant extracytoplasmic polypeptides (Strauch and Beckwith 1988). Although previous studies have estimated that ¢~ directs transcrip-

1Present address: Department of Molecular and Cellular Biology, Har- vard University, Cambridge, Massachusetts 02138 USA. 2Corresponding author. E-MAIL [email protected]; FAX (609) 258-6175.

tion of at least 11 genes (Raina et al. 1995; Rouvihre et al. 1995), only three loci are presently known to be transcribed by cE: (1) degP; (2) rpoH, encoding ¢32; and (3) rpoE, encoding ¢E itself (Erickson and Gross 1989; Wang and Kaguni 1989; Raina et al. 1995; Rouvihre et al. 1995).

¢~"s regulation of the periplasmic protease, DegP, sup- ports the view that cE is involved in controlling aspects of extracytoplasmic physiology in E. coli. However, DegP is presently the only extracytoplasmic protein whose synthesis is known to be controlled by this c-factor. To clarify the primary function of cE, more members of its regulon must be idenitified.

In a similar fashion to ¢F., the activity of the Cpx two- component signal transduction system is linked to the physiology of the bacterial envelope. For example, the activity of the two-component inner-membrane sensor, CpxA, is stimulated by overproduction of the outer- membrane lipoprotein, NlpE (Danese et al. 1995; Snyder et al. 1995). CpxA responds to this stimulus by communicating with its cognate response regulator, CpxR, which ultimately increases transcription of the degP locus (Danese et al. 1995; Raina et al. 1995).

In addition to being activated by extracytoplasmic protein-mediated signals, we have shown previously that activation of the Cpx system can combat the toxicities

GENES & DEVELOPMENT 11:1183-1193 © 1997 by Cold Spring Harbor Laboratory Press ISSN 0890-9369/97 $5.00 1183

Cold Spring Harbor Laboratory Press on October 6, 2020 - Published by genesdev.cshlp.orgDownloaded from

http://genesdev.cshlp.org/

http://www.cshlpress.com

Danese and Silhavy

conferred by the synthesis of certain mutan t envelope proteins (Cosma et al. 1995; Snyder et al. 1995). The Cpx system performs this stress-combative function, in part, by s t imulat ing the synthesis of the periplasmic protease, DegP. However, the Cpx system can partially combat these extracytoplasmic protein stresses even in the absence of DegP (Cosma et al. 1995; Snyder et al. 1995). Therefore, the Cpx pathway must control at least one other factor that can combat envelope protein toxicities.

The results described above indicate that although (rE and the Cpx proteins are influenced by and affect extracytoplasmic events, the precise functions of the (rE and Cpx systems are not f i rmly established. To further our understanding of the roles of both (rE and Cpx, we sought to identify and characterize new members of each of these regulons.

We have found that the Cpx and (rE systems each control the synthesis of periplasmic enzymes that can aid in protein folding. Because the Cpx and (rE systems also control the synthesis of the periplasmic protease, DegP, the results described in this study suggest that the primary functions of these two regulons may be to mediate protein folding and protein turnover wi th in the bacterial envelope.

Results

Activation of the o ~ and Cpx regulons alters the profile of periplasmic proteins

Because the (rE and Cpx systems exert their effects on the bacterial envelope, it seemed reasonable to assume that some members of their respective regulons would be found in the periplasm of E. coil Accordingly, we screened for periplasmic proteins whose synthesis could be activated by (r E or by the Cpx signal transduction system.

Figure la shows a Coomassie-stained, steady-state profile of periplasmic proteins from strain CLC198 (MC4100, degP::TnlO) transformed wi th either (1) a control plasmid (pBR322, lane 1) or (2) a plasmid that overproduces (rE (pND12, lane 2). The most striking differ- ence between these two lanes is the increased intensi ty of a band that migrates in the 32-kD size range (Fig. la, cf. lanes 1 and 2). Note that CLC198 contains a deg- P::TnlO mutation, which was uti l ized to mitigate any effects that increased proteolysis would have on the profile of periplasmic proteins during these experiments.

In a s imilar fashion, Figure lb shows the Coomassie- stained, steady-state profile of periplasmic proteins from strains SP779 [MC4100, KRS88(degP-lacZ), ara74::cam, zab::TnlO] and SP781 (SP779, cpxR::spc) transformed wi th either (1) a control plasmid (pBAD18, lanes 1,3) or (2) a plasmid that overproduces the outer-membrane lipoprotein, NlpE (pND18, lanes 2,4). Because overproduction of NlpE activates the Cpx signal transduction system (Danese et al. 1995; Snyder et al. 1995), the strain used to prepare the periplasmic extract shown in lane 2 of Figure lb possesses an activated Cpx signal transduction system. Lanes 3 and 4 of Figure lb show extracts

a

4 b

Lane 1 2 Lane

Z

1 2

~ z R : : s ~

Z m~

3 4

Figure 1. Activation of the cr E and Cpx regulons alters the profile of periplasmic proteins. (a) Activation of the cr E regulon increases the intensity of a band migrating in the 32-kD size range. Periplasmic extracts were prepared from CLC198 (MC4100, degP::TnlO)transformed with pBR322 (control for pND12) (lane I) or pND12 (overexpresses rpoE) (lane 2). The 32-kD band whose intensity rises from lane 1 to lane 2 is marked with an arrowhead. (b) Activation of the Cpx two-component signal transduction pathway increases the intensity of a band migrating in the 23-kD size range. Periplasmic extracts in lanes 1 and 2 were prepared from SP779 [MC4100, KRS88(degP- lacZ), ara 74::cam, zab::Tnl 0] transformed with either pBAD 18 (control for pND18)(lane 1) or pND18 (overexpresses nlpE) (lane 2). Periplasmic extracts in lanes 3 and 4 were prepared from SP781 (SP779, cpxR::spc) transformed with either pBAD 18 (lane 3) or pND18 (lane 4). The 23-kD band whose intensity rises from lane 1 to lane 2 is marked with an arrowhead. Strains used to generate the protein extracts depicted in a were grown in Luria broth with ampicillin. Strains used to generate protein extracts depicted in b were grown in Luria broth containing ampicillin and 0.4% L-arabinose.

from transformant derivatives of SP781, which contain a cpxR null mutat ion. Thus, these lanes serve as controls to help in determining whether any changes between the protein profiles of lanes 1 and 2 are actually dependent on the Cpx pathway. Figure 1 b shows a band of -23-kD in size whose intensi ty rises from lane 1 to lane 2. No change in the intensi ty of this band is observed when comparing lanes 3 and 4, indicating that the increased intensi ty of this band is dependent on CpxR.

Amino-terminal sequencing of the 32- and 23-kD bands

The 32- and 23-kD bands shown in Figure 1 were promising candidates for proteins that could be regulated by (r E and Cpx, respectively. Accordingly, the identi ty of the first 11 amino acid residues from the amino terminus of

1184 G E N E S & DEVELOPMENT




(r E a n d C p x c o n t r o l s y n t h e s i s o f p r o t e i n - f o l d i n g e n z y m e s

each protein was determined by Edman degradation (see Materials and Methods). The sequence determined for the 32-kD band (AEAAKPATAAD) corresponds to residues 26-36 of FkpA, a peptidyl-prolyl cis/trans isomerase identified by Home and Young (1995). Missiakas et al. (1996) have provided evidence recently suggesting that FkpA performs a protein-folding function within the extracytoplasmic compartments of E. coli. The sequence determined for the 23-kD band (AQYEDGKQYTT) corresponds to residues 20-30 of DsbA, a periplasmic protein required for efficient disulfide bond formation in E. coli (Bardwell et al. 1991; Kamitani et al. 1992).

The amino-terminal sequence determined for FkpA corresponds to the predicted amino terminus that would be generated after a signal-sequence-cleavage event. Taken together, the sequencing and fractionation of FkpA demonstrate that it is a periplasmic protein with a functional signal sequence. Note that the sequence determined for DsbA also corresponds to the signal-sequence processed form of the protein.

Transcriptional regulation of fkpA

There are several possible explanations for the increased amount of FkpA found in periplasmic extracts of strains overproducing (r~. However, because ~r z is involved in transcriptional initiation, the simplest model posits that increased levels of (rE concomitantly increase fkpA transcription.

Accordingly, we constructed an fkpA-lacZ operon fusion to determine whether the (r~-overproducing plasmid affected fkpA transcription. This fusion was recombined onto a X phage and was placed in single copy at the attB locus on the E. coli chromosome (see Materials and

Methods). f3-Galactosidase activities were determined from derivatives of SP887 [MC4100, hRS88(fkpA-Iacz)] that were transformed with either (1) the pBR322 control plasmid (Fig. 2a, lane 1) or (2)pND12, which overproduces (r E (Fig. 2a, lane 2). Figure 2a illustrates that the o-E-overproducing plasmid stimulates fkpA transcription approximately sevenfold when compared with its control strain. Thus, (rE can stimulate fkpA transcription.

fkpA transcription is affected by fluctuations in o ~ activity

To determine the extent of (rZ'S influence on fkpA transcription, we quantified the amount of fkpA transcription that is generated during situations in which (r E activity is altered by extracytoplasmic events.

For example, Mecsas and coworkers (1993) demonstrated that overproduction of the outer-membrane protein, OmpX, stimulates (rE activity approximately fourfold. We were therefore interested in determining whether fkpA transcription would also be induced by overproduction of OmpX. We measured the amount of fkpA-lacZ transcription generated from SP887 [MC4100, hRS88(fkpA-IacZ)] that contained either (1)a control plasmid, pBR322 (Fig. 2b, lane 1) or (2)pJE100, a plasmid that overproduces OmpX (Mecsas et al. 1993). Compari- son of lanes 1 and 2 of Figure 2b shows that overproduction of OmpX stimulates fkpA-IacZ transcription approximately twofold when compared with a control strain. Thus, overproduction of OmpX, which stimulates (rE activity, also stimulates fkpA transcription, albeit to a lesser extent than is observed with degP transcription.

The surA null mutation impairs the assembly of the porins LamB, OmpF, and OmpC, and as a result of this

a 4000

3 0 0 0

<

2 0 0 0

1000

o

b 400 -

3 0 0 -

2 0 0 -

C 2 5 0 0 -

2000 -

1500 -

1000 -

100t 500 i i 0

"G~ degP ~ ~ ~ ~ surA

t o t o

500 -

T

!!! Lane 1 2 Lane 1 2 Lane 1 2 3 4

Figure 2. Overproduction of (r E stimulates transcription of fkpA-lacZ. (a) ~-Galactosi- dase activities of SP887 [MC4100, hRS88(fkpA-IacZ)] transformed with pBR322 (control for pND12) (lane 1) or pND12 (overexpresses rpoE) (lane 2) were assayed. The rpoE overexpressing strain stimulates transcription of fkpA-lacZ approximately sevenfold over that of the control strain. (b) Overproduction of the outer- membrane protein, OmpX, stimulates fk- pA-IacZ transcription. ~-Galactosidase activities of SP887 [MC4100, XRS88(fkpA- lacZ)] transformed with either pBR322 (control for pJE100)(lane/)or pJE100 (overproduces OmpX) (lane 2) were determined. The OmpX overproducing plasmid increases fkpA-lacZ transcription approximately twofold when compared with the

control strain. (c) The surA- degP- double mutant stimulates fkpA-lacZ transcription. ~3-Galactosidase activities of SP887 (lane 1), SP940 (SP887, degP::TnlO) (lane 2), SP921 (SP887, surA::kan) (lane 3) and SP942 (SP921, degP::TnlO) (lane 4) were determined. The degP::TnlO mutation alone does not stimulate fkpA transcription (cf. lanes 1 and 2). In contrast, the surA null mutation alone stimulates fkpA transcription 1.4-fold, whereas the surA- degP- double mutant stimulates fkpA-lacZ transcription -4-fold when compared with the control strain. Strains were grown in Luria broth (with anapicillin when needed), and all procedures were performed as described in Materials and Methods.

G E N E S & D E V E L O P M E N T 1 1 8 5




Danese and Silhavy

assembly defect, the muta t ion also increases ¢~ activity approximately fivefold (Lazar and Kolter 1996; Missiakas et al. 1996; data not shown), surA specifies a periplasmic peptidyl-prolyl cis/trans isomerase, and it is believed that the SurA protein directly catalyzes a step(s) in the folding of the porins ment ioned above (Lazar and Kolter 1995; Missiakas et al. 1996).

The porin assembly defect conferred by the surA null muta t ion is aggravated by the absence of the DegP protease. Although the degP null muta t ion does not st imulate ¢r activity on its own, when it is introduced into a surA null strain, ¢E activity rises 10-fold (data not shown).

We were interested in determining the extent of fkpA transcriptional induction under these circumstances as well. Specifically, we quantified the transcription generated from the fkpA-lacZ fusion in either (1) a wild-type background (Fig. 2c, lane 1); (2)a degP- background (Fig. 2c, lane 2); (3) a surA- background (Fig. 2c, lane 3), or (4) a surA- degP- double-mutant background (Fig. 2c, lane 4). As expected, the degP null muta t ion has no effect on fkpA transcription. In contrast, the surA null increases fkpA transcription 1.4-fold when compared with the wild-type strain (Fig. 2c, cf. lanes 1 and 3). The surA- degP- double mutan t displays the largest induction of fkpA transcription at fourfold over that of the wild-type strain (Fig. 2c, cf. lanes 1 and 4). Again, these results indicate that extracytoplasmic defects that increase crr activity also s t imulate fkpA transcription.

A o ~ promoter stimulates fkpA transcription

The transcriptional induction of fkpA by overproduction of crr as well as by st imuli that activate ¢~ suggests that fkpA transcription is controlled, at least in part, by a Cr promoter. An analysis of the noncoding upstream sequence of fkpA highlights a putative ~r promoter (Fig. 3a). Nucleotides 342-347 of the published fkpA sequence (Home and Young 1995) contain 4 of the 6 consensus nucleotides for a (rE -35 promoter site (aAACTa). This site is followed by a consensus 16-nucleotide-long spacer region that contains a string of 5 adenine nucleotides that is also characteristic of cr E promoters (Lipinska et al. 1988; Erickson and Gross 1989; Raina et al. 1995; Rou- vihre et al. 1995). This spacer region is followed by a -10 site that possesses 5 of 5 consensus nucleotides (TC- TGA).

In light of this putat ive cr E promoter, we wanted to determine the start site of the fkpA transcripts that are induced upon overproduction of crE. To this end, we performed S1 nuclease protection assays using RNA prepared from a strain that was transformed with either (1) pBR322 (control for pND12) or (2) pND12 (overproduces erE). Figure 3b shows that the transcripts induced by overproduction of cr E begin at nucleotides 370, 371, and 372 of the published fkpA sequence, which correspond to positions of +4, +5, and +6 wi th respect to the ¢E -10 consensus promoter site (see Fig. 3a) (Home and Young 1995).

Although this init iation site is relatively close to the

a flcpA 6 ' ' soo

a40 C G ~ A T T T A A A c[AAA AAIG A ~TC TG AIA AA'FAGA 375

b

oE overproduced - + Lane C T A G 1 2

5'

~ tr.,a~nt

fkpA I

]kpA 2 3'

Figure 3. Overproduction of O "E stimulates transcription from a site downstream of a ¢E-like promoter. (a) The fkpA coding sequence is shown as a shaded rectangle with a scale (in nucleotides) below. A portion of the upstream noncoding sequence of fkpA is shown below the scale. This nucleotide sequence corresponds to positions 340-375 in the published sequence of Home and Young (1995). The regions comprising the compo- nents of the erE-like promoter are enclosed in rectangles, and consensus nucleotides are shown in boldface. The initiation sites of transcripts induced by overproduction of CE (see b) are underlined. (b) Lanes 1 and 2 show the fkpA transcription start sites from strain SP887 [MC4100, hRS88(fkpA-lacZ)] transformed with either a control plasmid, pBR322 (lane 1), or the cr E overproducing plasmid pND12 (lane 2). A digested fragment corresponding to the 5' end of the spc operon is also noted by the arrow marked "spc fragment." This fragment serves as an internal loading control. Two groups of fkpA transcripts are shown (fkpA-1 and fkpA-2). The fkpA-1 arrow shows protected DNA fragments that correspond to transcripts that initiate at positions 371, 372, and 373 of the fkpA sequence (Home and Young 1995). These positions are underlined in a, and they correspond to positions +4, +5, and +6 relative to the ¢E-10 consensus promoter site shown in a. Note that the fkpA-1 fragments are only present in lane 2 (gE overproduced). The fkpA-2 arrow shows a protected DNA fragment that corresponds to a transcript that initiates at position 401 of the fkpA sequence (Home and Young 1995). Note that the fkpA-2 fragment is only present in lane 1 (¢E not overproduced). The lanes labeled C, T, A, and G indicate dideoxy sequencing reactions initiated by the same oligonucleotide used to label the S1 probe (Fkpalac3). The 5' to 3' direction of the nontemplate nucleotide sequence of fkpA is from the top of the b to the bottom, as shown by the 5' --~ 3' arrow. RNA was prepared from strains grown at 37°C in Luria broth containing 125 pg/ml of ampicillin.

cr E -10 site [the strongest ¢E transcripts begin at positions +8 and +9 (Raina et al. 1995; Rouvihre et al. 1995)],

1186 GENES & DEVELOPMENT




other E. coli promoters also initiate transcription at these early sites (Harley and Reynolds 1987). Moreover, the results presented in Figure 3b may help to explain the attenuated induction of fkpA transcription that is observed upon activation of ~r E.

Recall that situations in which cr E induces activity 4-fold will stimulate fkpA transcription only by 1.4- to 2-fold (Fig. 2b). There are at least three possible explanations for this observation. First, the position of the transcriptional initiation sites relative to the cr E -10 site may diminish the ability of RNA polymerases containing cr E to transcribe this promoter. Second, the cr E -35 site of fkpA does not possess 100% identity with the consensus -35 site (see Figs. 3a and 4). The crE-regulated rpoEp2 promoter, which contains a nonconsensus -10 site and a transcriptional initiation site of + 10, also displays an attenuated response to activation by CE (Rouvihre et al. 1995). Third, lane 1 of Figure 3b shows that when ¢E is not overproduced, a transcriptional initiation site is located at position 401 of the fkpA sequence. When ¢E is overproduced, this transcript disappears (Fig. 3b, lane 2). Thus, the total amount of transcriptional induction that is observed with the fkpA-lacZ fusion is attenuated because ~r E overproduction stimulates transcriptional initiation at nucleotides 370-372, at the expense of transcription initiation at nucleotide 401.

We have also noted that the rpoE null mutation (which eliminates CE synthesis) does not reduce transcription of fkpA-lacZ (data not shown). However, this is not sur- prising in light of the results presented in Figure 3b. When ~r ~ is not activated (i.e., nonstressing conditions), the majority of transcription generated from fkpA initiates at position 401. It is only upon activation of ~r E that we observe transcriptional initiation at sites 370-372.

Taken together, the results of Figure 3b show that fkpA possesses a ¢E-activatable promoter that has the standard features recognized by crE. Figure 4 shows this fkpA promoter aligned with the three CE promoters of E. coli characterized previously.

Transcriptional regulation of dsbA

In a similar fashion to CE's regulation of FkpA, there are several possible explanations for the increased amount of DsbA found in periplasmic extracts of strains possessing an activated Cpx signal transduction system. However, because the Cpx pathway controls the transcription of degP (Danese et al. 1995; Raina et al. 1995), and because CpxR is homologous to the OmpR subfamily of two-

ar E and Cpx control synthesis of protein-folding enzymes

component transcription factors (Dong et al. 1993), the simplest interpretation of the results presented in Figure lb is that activation of the Cpx pathway stimulates dsbA transcription.

To test the possibility that the Cpx pathway could activate transcription from the dsbA locus, we created an orfA-dsbA-lacZ operon fusion. This fusion was recombined onto a X phage and was placed in single copy at the attB locus on the E. coli chromosome (see Materials and Methods).

The incorporation of orfA into the operon fusion was necessitated because the dsbA gene is situated in an operon with an upstream gene, orfA, of unknown function (Fig. 5a). Transcription of dsbA is directed from two promoters, each of which contributes to approximately one- half of DsbA synthesis. The first promoter is situated within the orfA (pdsbA) coding sequence, whereas the second is positioned upstream of orfA (porfA-dsbA). This latter promoter cotranscribes both orfA and dsbA (Fig. 5a)(Belin and Boquet 1994).

To ensure that all of the transcripts used to synthesize DsbA were represented in the operon fusion, we fused the promoter region of orfA, the entire orfA coding sequence (which contains the first dsbA promoter), and the first 86 nucleotides of the dsbA coding sequence to the lac operon, creating porfA-dsbA-lacZYA (Fig. 5a).

We first wanted to determine whether dsbA transcription could be stimulated by the cpxA24 allele, which possesses a mutation that hyperactivates the Cpx pathway and stimulates degP transcription approximately eightfold (Danese et al. 1995). Accordingly, we introduced the cpxA24 mutation into strain SP969 [MC4100, KRS88(porfA-dsbA-IacZ)], generating strain SP971 (SP969, cpxA24). Comparison of lanes 1 and 2 in Figure 5b shows that the cpxA24 mutation stimulates porfA- dsbA-lacZ transcription sixfold, indicating that activation of the Cpx pathway stimulates dsbA transcription.

The transcriptional induction of dsbA is also observed when the outer membrane lipoprotein, NlpE, is overproduced. Overproduction of NlpE has previously been shown to activate the wild-type Cpx pathway and stimulate degP transcription (Danese et al. 1995; Snyder et al. 1995). Figure 5c shows the results of f~-galactosidase assays performed on strains SP994 [MC4100, kRS88(porfA- dsbA-lacZ), ara74::cam, zab::TnlO], SP995 (SP994, cpxA::cam), and SP996 (SP994, cpxR::spc) transformed with either (1) pBAD 18 (control for pND 18) or (2) pND 18 (overproduces NlpE). Comparison of lanes 1 and 2 in Fig- ure 5c shows that NlpE overproduction stimulates

rpoHp3 T T

degP T C

rpoEp2 C G

. t ~ c c

-35

G A A CT T

G A A C T T

G A A C T T

A A A C T A

-I0

GTGGAT[AAAA]TC A C G G T C T G A r A A A A CA

CAGG C T A T ~ A A A ] C G A A T C T G A ~ G A A C A C

]'A C ~ A A A A ~ E G A G A C A C T C T A A 2C C T TT GC

ATT TAAAC~AAAA A ~ A G T C T G A k A A T A G A T

Figure 4. Alignment of ~E-activatable promoters. The -35, -10, and poly(A) regions are boxed. The transcription start sites are underlined. See text for details.

GENES & DEVELOPMENT 1187




Danese and Silhavy

lacZYA

porfA-dsbA Pr~ bA I L ~i~;~,~ i~i~i~:!~:~:~i~;~!~!~i~!~!~l~] _'l~...~.!~.~!~i~N~ l

, orfA dsbA '1000 ' '2000 . . . . 3000

800 g~

"~ 600 <

.~ 4oo

0

200

c~ 0

bo

Lane 1 2

600 c

.~ '~ 500

,,I,,o t~

'< 4o0

~ 3o0

0

~ 2oo

100

0 NlpE

Overproduced cpxA cpxR

Lane 1 2 3 4 5 6

Figure 5.Activation of the Cpx pathway stimulates transcription of the porfA-dsbA-lacZ operon fusion. (a) The orfA-dsbA operon and the porfA-dsbA-lacZ operon fusion. The orfA and dsbA open reading frames are shown as shaded rectangles. The two promoters that transcribe dsbA (porfA-dsbA and pdsbA) are depicted as arrows emanating from their respective initiation sites. The genomic DNA used to create the porfA-dsbA-lacZ operon fusion is shown as a thin line fused to the lacZ operon above the orfA and dsbA coding sequences. A 3000-nucleotide-long scale is shown below the operon for reference. (b) A hyperactive cpxA allele stimulates transcription of porfA-dsbA-lacZ. [5-Galactosidase activities from SP969 [MC4100, KRS88(porfA-dsbA-lacZ)] and SP971 (SP969, cpxA24)were determined. The cpxA24 mutation stimulates dsbA-lacZ transcription approximately sixfold. (c) Overproduction of NlpE stimulates porfA-dsbA-lacZ transcription by the Cpx pathway. Lanes 1, 3, and 5 show ~-galactosidase activity of strains transformed with pBAD18 (control for pND18). Lanes 2, 4, and 6 show the [~-galactosidase activity of strains transformed with pND18 (overexpresses nlpE). (Lanes 1,2)SP994 [MC4100, )~RS88(porfA-dsbA-lacZ), ara74::cam, zab::TnlO]. (Lanes 3,4)SP995 (SP994, cpxA::carn). (Lanes 5,6) SP996 (SP994, cpxR::spc). All strains were grown in Luria broth (containing 0.4% L-arabinose and ampicillin when needed).

porfA-dsbA-lacZ transcription 5.3-fold. Only a minor s t imulatory effect is observed in the cpxA- and cpxR- cpxA- backgrounds (Fig. 5c, cf. lanes 1 and 2 with lanes 3 and 4; 5 and 6), indicating that overproduction of NlpE st imulates dsbA transcription by activating the wild- type CpxA protein.

Activation of the Cpx pathway stimulates dsbA transcription from the promoter upstream of the orfA locus

Because dsbA transcription originates from two promoters, one wi th in the orfA coding sequence and one that also cotranscribes orfA (Fig. 5a), we were interested in determining which of these two promoters was uti l ized by the activated Cpx pathway. To address this issue, we used S1 nuclease protection assays to quantify the amount of transcription generated from the porfA-dsbA promoter and the pdsbA promoter using RNA prepared from strains that contain either an nlpE overexpressing plasmid (pND18) or a control plasmid (pBAD18). Recall that overproduction of NlpE activates dsbA transcription in a CpxA-dependent fashion (Fig. 5c).

The pdsbA promoter is unaffected by overproduction of NlpE (data not shown). In contrast, Figure 6a shows that transcription from the porfA-dsbA promoter is

s t imulated by overproduction of NlpE. Also, the primary transcriptional ini t ia t ion site induced by overproduction of NlpE corresponds to the + 1 site described for porfA- dsbA by Belin and Boquet (1994) (see Fig. 6a,b). There- fore, overproduction of NlpE increases the synthesis of DsbA by s t imulat ing transcription from the porfA-dsbA promoter.

o ~ and Cpx do not coregulate transcription of fkpA and dsbA

Previous studies have indicated that the Cpx signal transduction system and cr E jointly control the transcription of the degP locus (Danese et al. 1995; Raina et al. 1995). Accordingly, we were interested in determining whether the transcription of any of the other members of the Cpx and cr E regulons were also regulated jointly. To address this issue, we determined whether the overproduction of (r E could activate dsbA transcription, and we also determined whether activation of the Cpx pathway (by overproduction of NlpE) could activate fkpA transcription. In the former case, overproduction of cr E does not activate dsbA transcription, and in the latter case, overproduction of NlpE does not s t imulate fkpA transcription (data not shown). These results indicate that





crr and Cpx control synthesis of protein-folding enzymes

at I~E ocerproduced - +

porfA-dsbA

spc fragment

Lane 1 2

b 1014 A A A GCT TGTA A GCGGCGCCA CCAA A A TCA TC

TGAAA TGA[TA TCCT]TCGTCA TTCGT lff~ -1

] ,L

........... "~ ~ i i ~ .................. ~ '~;~'~i

, o~A d~A . . . . 1 ~ ' 'Z000 . . . . 30fJ0

Figure 6. Activation of the Cpx pathway by overproduction of NlpE stimulates transcription from the porfA-dsbA promoter. (a) Lanes 1 and 2 show major porfA-dsbA transcription start sites as well as minor, nearby start sites from strain SP994 [MC4100, XRS88(porfA-dsbA-lacZ), ara74::cam, zab::TnlO] transformed with either a control plasmid, pND18 (lane 1), or the nIpE overexpressing plasmid pND18 (lane 2). An $1 nuclease-protected fragment corresponding to the 5' end of the spc operon is also noted by the arrow marked "spc fragment." This fragment serves as an internal loading control. The arrow marked "porfA-dsbA" shows protected DNA fragments that correspond to transcripts that initiate at positions 1066 and 1067 of the published orfA-dsbA sequence (Belin and Boquet 1994). Position 1067 is the transcription initiation site first described by Belin and Boquet (1994), and this position is underlined in b. (b) The porfA-dsbA promoter. The nucleotide sequence surrounding the porfA-dsbA promoter is shown, spanning from nucleotide 1014 to 1070 of the published sequence (Belin and Boquet 1994). The putative -10 site of this promoter is boxed, and the major transcription initiation site is underlined. The orfA-dsbA operon is shown below the nucleotide sequence, and a 3000-nucleotide-long scale is also shown for reference.

although Cpx and cr E intersect in their regulation of degP, their regulons do not completely overlap.

D i s c u s s i o n

The control of protein-folding agents within the bacterial envelope

The results presented here provide the first evidence that the cr E and Cpx regulons not only contain proteases like DegP, but that they also contain periplasmic enzymes that can engage in protein-folding activities. This result implies that there is a second envelope-protein stress-

combative tool at the disposal of E. coli--specifically, protein-folding agents. Thus, much like ~r 32 controls the synthesis of a host of cytoplasmic proteases and molecular chaperones (Gross 1996), the ~r ~ and Cpx regulons may perform a complementary function, mediating protein folding and protein turnover within the bacterial envelope.

FkpA and the c a regulon

Missiakas et al. (1996) have recently provided evidence that high-level synthesis of FkpA can suppress extracytoplasmic stresses, such as the accumulation of unfolded periplasmic and outer-membrane proteins. These au- thors have also demonstrated that FkpA is a peptidyl- prolyl cis/trans isomerase, suggesting that the function of this protein is to facilitate the folding of other extracytoplasmic proteins.

The results presented here provide a satisfying complementary analysis to that performed by Missiakas et al. (1996). Specifically, we have shown that both the overproduction of cE and the creation of extracytoplasmic stresses that stimulate cr E activity will increase the synthesis of FkpA. This increased synthesis is mediated by a crE-activatable promoter that shares the common features of the other known ¢E-regulated promoters of E. coli (Fig. 4). Taken together, these results suggest that fkpA is the newest member of the ~r E regulon.

Cpx and DsbA

In addition, the results presented here suggest that the Cpx pathway is also involved in mediating protein-folding functions within the bacterial envelope. Specifically, DsbA synthesis is increased by activation of the Cpx pathway. Interestingly, this study indicates that the Cpx pathway also stimulates transcription of orfA, the gene upstream of dsbA (Fig. 5a). However, the function of this gene is presently unknown.

The Cpx-mediated stimulation of DsbA synthesis is intriguing for several reasons: First, we have shown previously that activation of the Cpx pathway can combat extracytoplasmic protein-mediated toxicities (Cosma et al. 1995; Snyder et al. 1995). The activated Cpx system performs this function, in part, by stimulating the synthesis of DegP. However, the activated Cpx system can still partially suppress these stresses even in the absence of DegP, indicating that there are other stress-combative members of the Cpx regulon. DsbA is a promising can- didate for such a factor. However, Snyder and Silhavy (1995) have demonstrated that the dsbA null mutation does not impede the ability of the Cpx pathway to ame- liorate the periplasmic toxicity associated with the Lam- B-LacZ-PhoA fusion protein. Although this does not preclude the possibility that the Cpx pathway can utilize DsbA in other stress-combative situations, the result does imply that there is still at least one other uniden- tified factor utilized by the Cpx regulon to combat extracytoplasmic protein-mediated toxicities.

Second, from the results presented, it is not clear





Danese and Silhavy

whether transcription of any of the other Dsb proteins (DsbB, DsbC, and DsbD) can also be s t imulated by the Cpx pathway (Bardwell et al. 1993; Dailey and Berg 1993; Missiakas et al. 1993, 1994, 1995). Because this study has only searched for periplasmic Cpx-inducible proteins, inner-membrane Dsb proteins (DsbB and DsbD) would not have been identified by this approach. Accordingly, an analysis of membrane proteins in Cpx-activated strains could be informative in identifying the remaining members of the Cpx regulon.

Three classes of o ~ and Cpx regulatory targets

Figure 7 shows a list of Cpx-regulated and erE-regulated loci. There are now three types of genes that are controlled by either the Cpx or cr v regulons: (1) those that are controlled solely by Cpx (dsbA), (2) those that are controlled solely by cr E (rpoHp3, rpoEp2, fkpA), and (3) those that are jointly controlled by both Cpx and ~r E (degP) (Lipinska et al. 1988; Erickson and Gross 1989; Danese et al. 1995; Raina et al. 1995).

The simplest explanation for the interactions depicted in Figure 7 is that all interactions are direct. For example, CpxR would bind upstream of its regulatory targets (degP, orfA-dsbA) and s t imulate transcription from the promoters of these genes. Pogliano et al. (this issue) have demonstrated footprinting of CpxR at the degP and dsbA loci, lending support to this model.

In sum, the results presented in this paper and in previous studies indicate that the Cpx two-component signal transduction system and the heat shock-inducible or-factor ¢E both control the synthesis of a periplasmic protease (DegP) and periplasmic enzymes that are ca- pable of performing protein-folding functions (Lipinska et al. 1988; Erickson and Gross 1989; Danese et al. 1995; Raina et al. 1995). Given these two classes of proteins, it seems l ikely that the primary functions of the Cpx and

~E systems are to monitor and mediate protein-folding and protein-turnover functions wi th in the extracytoplasmic compartments of E. coli. We would not be surprised if many of the remaining unidentif ied members of the Cpx and o~ E regulons fall into these two classes of proteins.

Mater ia l s and m e t h o d s

Media

Media were prepared as described (Silhavy et al. 1984). Liquid cultures were grown in Luria broth. The final concentration of ampicillin used in the growth media was 50 lag/ml. Standard microbiological techniques were used for strain construction and bacterial growth (Silhavy et al. 1984).

Strains and phage

All strains are derivatives of MC4100 (Silhavy et al. 1984). The genotypes of all strains used in this study are given in the leg- ends to Figures 1, 2, 3, 5, and 6 and in the text. The cpxA24 mutation (Fig. 5b) was moved by P1 transduction, as described previously (Danese et al. 1995).

The ara 74::cam mutation (gift of Leslie Pratt, Harvard Medi- cal School, Boston, MA) confers upon MC4100 an arabinose- resistant and arabinose-minus phenotype (MC4100 is normally sensitive to growth in the presence of arabinose). The ara74::cam mutation was introduced into strains when the arabinose-dependent induction of NlpE synthesis was required (Figs. lb, 5c, and 6). The ara74::cam mutation was moved with a linked TnlO insertion (zab::TnlO), and the resulting transduc- tants were scored for the ability to grow in the presence of arabinose.

The zab::TnlO, cpxA::cam, cpxR::spc, degP::TnlO, and surA- ::kan (kind gift of Sara Lazar and Roberto Kolter) mutations were moved by P1 transduction, selecting for resistance to the appropriate antibiotic [(TnI01 tetracycline; (cam)chlorampheni- col; (spc) spectinomycin; (kan) kanamycin]. The degP::TnlO mutation was assayed further for the conferral of temperature- sensitive growth at 42°C.

Figure 7. A model of the cr and Cpx regulons. The Cpx two-component signal transduction system is shown as being activated by the outer-membrane lipoprotein, NlpE. The inner-membrane sensor, CpxA, is shown phosphorylating its cognate response regulator, CpxR, whereupon the activated CpxR-phosphate molecule stimulates transcription of the degP and dsbA genes. The DegP and DsbA proteins are shown in the periplasm, performing proteolytic and protein-folding functions, respectively. An unknown signal transduction pathway [depicted as a question mark (?)] is shown monitoring the level of an outer-membrane protein and stimulating crr activity. ¢E is shown activating transcription of fkpA, degP, rpoHp3, and rpoEp2. The FkpA protein is also shown in the periplasm, facilitating the interconver- sion of cis and trans isomers of prolyl-containing peptides.

. . . . . . . . mmmlmm

M •

ATP

degP

dsbA jkpA rpoHp3 rpoEp2





uE and Cpx control synthesis of protein-folding enzymes

ARS88 has been described (Simons et al. 1987). Lysogeniza- tion of ARS88(fkpA-lacZ) and ARS88(porfA-dsbA-lacZ) was performed as described by Simons et al. (1987). All ARS88 operon fusions were shown to be located in single copy at the Aatt locus by P1 transduction.

Plasmid construction

All plasmids used in this study confer ampicillin resistance. pND12, which overproduces uE, was constructed as follows:

rpoE was amplified from the chromosome of MC4100 by PCR using the Rpoe3 (5'-CTATCCAGCGTGTCGACATCCAT- TAAAGCGG-3') and Rpoe5 (5'-GCATGACAAACAAAAACG- GATCCGTTACGGAAC-3') primers. SalI and BamHI restriction sites were incorporated into the Rpoe3 and Rpoe5 primers, respectively, to facilitate subcloning. The amplified DNA was then subcloned into the BamHI and SalI sites of pBR322, creating pND12. pND12 places rpoE under its own transcriptional control. The sequence of the pND12 insert was confirmed by dideoxynucleotide sequencing.

pJElOO overproduces the outer-membrane protein OmpX (Mecsas et al. 1993). The parent vector for pJElOO is pBR322.

pND18 expresses the nlpE locus under the control of the arabinose-inducible pBAD promoter (Danese et al. 1995). pBAD 18, the parent vector for pND 18, has been described (Guz- man et al. 1995).

Construction of XR S88(fkpA-lacZ)

The FkpA5 (5'-CTTCAATGGTGAATTCCTGAAAAG-3') and Fkpalac3 (5'-CAAAAGTGATTGGTGGATCCAGGGCAAC- 3') primers were used to amplify the promoter region and a portion of the fkpA open reading frame from the chromosome of MC4100. EcoRI and BamHI restriction sites were incorporated into the FkpA5 and Fkpalac3 primers, respectively, to facilitate subcloning. The amplified DNA was subcloned into the EcoRI and BamHI sites of pRS415, generating pND28. This amplified DNA includes nucleotides from position -286 with respect to the fkpA translational start site to position +55 with respect to this same site. The nucleotide sequence of the fkpA insert of pND28 was confirmed by dideoxynucleotide sequencing. The fkpA-lacZ fusion of pND28 was then recombined onto phage ARS88, and recombinants were used to lysogenize MC4100 as described (Simons et al. 1987).

Construction of hRS88fporfA-dsbA-lacZ)

The Dsba5 (5'-CGTCGTCATCGAATTCACCGATATCG- 3') and Dsba3 (5'-CAGCGCCAGCGAATTCTTTTTCATG-3') primers were used to amplify from the promoter region of orfA (the gene immediately upstream of dsbA) through to the begin- ning of the dsbA coding sequence. EcoRI restriction sites were incorporated into the Dsba5 and Dsba3 primers to facilitate subcloning. This DNA was amplified from the chromosome of MC4100 and subcloned into the EcoRI site of pRS415, generating pND3 1. The proper orientation of the insert was confirmed by restriction analysis. pND3 1 drives lac2 transcription from both promoters that are known to initiate transcription of dsbA, porfA, and pdsbA (Belin and Boquet 1994). The amplified DNA contains nucleotides from position -422 with respect to the translational start site of orfA to position +86 with respect to the start site of translation of dsbA. The porfA-dsbA fusion of pND31 was then recombined onto phage ARS88, and recombinants were used to lysogenize MC4100 as described (Simons et al. 1987).

p-Galactosidase assays

Cells were grown overnight in Luria broth. Cells were then subcultured (1:40) into 2 ml of the same meQa and grown to mid-log phase. p-Galactosidase activities were determined using a microtiter plate assay (Slauch and Silhavy 1991). p-Galac- tosidase activities are expressed as (UIA,,,) x lo3, where U = pmole of product formed per minute. Assays were performed on a minimum of four independent cultures of each strain, and the results were averaged to obtain the indicated activities. Error bars indicate the S.D. The absence of error bars indicates that the S.D. fell below the resolution limit of the graphing program.

Preparation of periplasmic protein extracts

All procedures were performed on ice, and all solutions were chilled on ice.

Periplasmic protein extracts were prepared as follows: Strains were grown in the appropriate media until they reached an OD,,, of -1 .O. One milliliter of each culture was harvested and resuspended in 250 pl of 0.2 M Tris-HC1 (pH 8.0). Two hundred and fifty microliters of 0.2 M Tris-HC1, 1 M sucrose (pH 8.0) was then added to each suspension, and 2.5 pl of 0.1 M EDTA was added subsequently along with 7.5 pl of lysozyme (4 mg/ml). Then, 500 pl of distilled water was added, and the cell suspension was incubated on ice for 2 min. Twenty microliters of 1 M

MgCl, was added to the cell suspension, and each mixture was incubated for 30 additional min on ice. After this incubation, each suspension was harvested at 14,000 rpm in a microcentri- fuge to pellet spheroplasts. The remaining supernatant contained the bulk of periplasmic proteins. The periplasmic proteins were precipitated from these supernatants with trichloro- acetic acid (TCA). The TCA-precipitated samples were resuspended in a milliliter volume equal to the initial OD,,, value/5. Twenty-microliter samples of these periplasmic protein extracts were resolved by polyacrylamide gel electrophoresis and subsequently stained with Coomassie brilliant blue (Sambrook et al. 1989).

Amino acid sequence analysis of FkpA and DsbA

Periplasmic protein extracts depicted in Figure 1 were subjected to SDS-polyacrylamide gel electrophoresis and transferred to Immobilon-P transfer membrane. The membrane was stained with Coomassie brilliant blue, and the bands marked with ar- rowheads in Figure 1 were excised. The identity of the first 11 residues from each band was determined by Edman Degradation by the Princeton University Synthesis/Sequencing Facility.

Preparation of E. coli RNA, S1 nuclease protection assays, and DNA sequencing

RNA was prepared from strains grown at 30°C in Luria broth as described by Barry et al. ( 1980). A 342-nucleotide-long fragment, spanning from position 833 to position 1175 of the published slyD-fkpA sequence (Horne and Young 1995)) was used to create a radioactive probe for S1 experiments depicted in Figure 3. This probe was used to determine the transcriptional start sites used to express the fkpA locus.

A 425-nucleotide-long fragment, spanning from position 931 to position 1356 of the published orfA-dsbA sequence (Belin and Boquet 1994), was used to create a radioactive probe for Sl experiments depicted in Figure 6. This probe was used to quantify the amount of transcription from the porfA promoter described by Belin and Boquet (1994).





Danese and Silhavy

A 270-nucleotide-long fragment spanning the pspc promoter was used to create a radioactive probe for S1 experiments. The pspc transcript serves as an internal loading control for quanti- fying changes in the amount of transcription from the fkpA and porfA promoters.

Each probe was phosphorylated with either [~-32p]ATP or [7-33P]ATP in the forward reaction as described (Sambrook et al. 1989). Sixty micrograms of total RNA was used in each S1 assay, and the assays were performed as described in Sambrook et al. (1989). The DNA sequence of fkpA and orfA-dsbA was determined as described previously (Russo et al. 1993). The fkpA and orfA-dsbA sequencing reactions and S1 nuclease samples were resolved on 6% polyacrylamide sequencing gels and ana- lyzed using the PhosphorImager ImageQuant (Molecular Dy- namics) analysis program.

A c k n o w l e d g m e n t s

We thank the members of the Silhavy laboratory, Jon Beckwith, Carol Gross, A. Simon Lynch, and Joseph Pogliano for helpful discussions. We also thank Jon Beckwith, Dominique Belin, Ed Lin, A. Simon Lynch, and Joseph Pogliano for communicating results prior to publication. We thank Christine Cosma for providing strain CLC 198, Sara Lazar and Roberto Kolter for providing the surA::kan mutation and Leslie Pratt for providing the ara74::cam mutation. P.N.D. gratefully acknowledges support from National Institutes of Health (NIH) training grant GM07388. This work was supported by a grant from NIH (GM34821) to T.J.S.

The publication costs of this article were defrayed in part by payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 USC section 1734 solely to indicate this fact.

R e f e r e n c e s

Bardwell, J.C., K. McGovern, and J. Beckwith. 1991. Identifica- tion of a protein required for disulfide bond formation m vivo. Ceil 67: 581-589.

Bardwell, J.C., J.O. Lee, G. Jander, N. Martin, D. Belin, and J. Beckwith. 1993. A pathway for disulfide bond formation in vivo. Proc. Natl. Acad. Sci. 90" 1038-1042.

Barry, G., C. Squires, and C.L. Squires. 1980. Attenuation and processing of RNA from the rplJL-rpoBC transcription unit of RNA polymerase. Proc. Natl. Acad. Sci. 77" 3331-3335.

Belin, P. and P.L. Boquet. 1994. The Escherichia coli dsbA gene is partly transcribed from the promoter of a weakly expressed upstream gene. Microbiology 140: 3337-3348.

Cosma, C.L., P.N. Danese, J.H. Carlson, T.J. Silhavy, and W.B. Snyder. 1995. Activation of the Cpx two-component signal transduction pathway in Escherichia coli suppresses envelope associated stresses. MoI. Microbiol. 18: 491-505•

Dailey, F.E. and H.C. Berg. i993. Mutants in disulfide bond formation that disrupt flagellar assembly in Escherichia coli. Proc. Natl. Acad. Sci. 90" 1043-1047.

Danese, P.N., W.B. Snyder, C.L. Cosma, L.J.B. Davis, and T.J. Silhavy. 1995. The Cpx two-component signal transduction pathway of Escherichia coli regulates transcription of the gene specifying the stress-inducible periplasmic protease, DegP. Genes & Dev. 9: 387-398.

Dong, J.S., S. Iuchi, H.S. Kwan, Z. Lu, and E.C.C. Lin. 1993. The deduced amino-acid sequence of the cloned cpxR gene suggests the protein is the cognate regulator for the membrane sensor, CpxA, in a two-component signal transduction system of Escherichia coli. Gene 136: 227-230.

Erickson, J.W. and C.A. Gross. 1989. Identification of the ¢E subunit of Escherichia coli RNA polymerase: A second al- ternate ~ factor involved in high-temperature gene expression. Genes & Dev. 3" 1462-1471.

Gross, C.A. 1996. Function and regulation of the heat shock proteins. In Escherichia coli and Salmonella, cellular and molecular biology, 2nd ed. (ed. F.C. Neidhardt), pp. 1382- 1399. ASM Press, Washington, DC.

Guzman, L.M., D. Belin, M.J. Carson, and J. Beckwith. 1995. Tight regulation, modulation, and high-level expression by vectors containing the arabinose pBAD promoter. J. Bacte- riol. 177: 4121-4130.

Harley, C.B. and R.P. Reynolds. 1987. Analysis of E. coli promoter sequences. Nucleic Acids Res. 15: 2343-2361.

Home, S.M. and K.D. Young. 1995. Escherichia coli and other species of the Enterobacteriaceae encode a protein similar to the family of Mip-like FK506-binding proteins. Arch. Micro- biol. 163: 357-365.

Kamitani, S., Y. Akiyama, and K. Ito. 1992. Identification and characterization of an Escherichia coli gene required for the formation of correctly folded alkaline phosphatase, a periplasmic enzyme. EMBO J. 11: 57-62.

Lazar, S.W. and R. Kolter. 1996. SurA assists the folding of £sch- erichia coli outer membrane proteins. J. Bacteriol. 178: 1770-1773.

Lipinska, B., S. Sharma, and C. Georgopoulos. 1988. Sequence analysis and regulation of the htrA gene of Escherichia coli: A eg2-independent mechanism of heat-inducible transcription. Nucleic Acids Res. 16" 10053-10067.

Lonetto, M., K.L. Brown, K.E. Rudd, and J.J. Buttner. 1994. Analysis of the Streptomyces coelicolor sigmaE gene reveals the existence of a subfamily of eubacterial RNA polymerase sigma factors involved in the regulation of extracytoplasmic functions. Proc. Natl. Acad. Sci. 91- 7573-7577.

Mecsas, J., P.E. Rouvihre, J.W. Erickson, T.J. Donohue, and C.A. Gross. 1993. The activity of cr E, an Escherichia coli heat- inducible ~-factor, is modulated by expression of outer membrane proteins. Genes & Dev. 7" 2618-2628.

Missiakas, D., C. Georgopoulos, and S. Raina. 1993. Identifica- tion and characterization of the Escherichia coli gene dsbB, whose product is involved in the formation of disulfide bonds in vivo. Proc. Natl. Acad. Sci. 90: 7084-7088.

• 1994. The Escherichia coli dsbC (xprA) gene encodes a periplasmic protein involved in disulfide bond formation. EMBO J. 13" 2013-2020.

Missiakas, D., F. Schwager, and S. Raina. 1995. Identification and characterization of a new disulfide isomerase-like protein (DsbD) in Escherichia coli. EMBO J. 14: 3414-3424.

Missiakas, D., J.-M. Betton, and S. Raina. 1996. New compo- nents of protein folding in extracytoplasmic compartments of Escherichia coli SurA, FkpA and Skp/OmpH. Mol. Micro- biol. 21: 871-884.

Pogliano, J., A.S. Lynch, D. Belin, E.C.C. Lin, and J. Beckwith. 1997. Regulation of Escherichia coli cell envelope proteins involved in protein folding and degradation by the Cpx two- component system. Genes & Dev. (this issue).

Raina, S., D. Missiakas, and C. Georgopoulos. 1995. The rpoE gene encoding the sigma E (sigma 24) heat shock sigma factor of Escherichia coli. EMBO J. 14: 1043-1055.

Rouvihre, P.E., A. de las Pefias, J. Mecsas, C.Z. Lu, K.E. Rudd, and C.A. Gross. 1995. rpoE, the gene encoding the second heat-shock sigma factor, ¢E, in Escherichia coil EMBO J. 14: 1032-1042.

Russo, F., J.M. Slauch, and T.J. Silhavy. 1993. Mutations that affect separate functions of OmpR, the phosphorylated regulator of porin transcription in Escherichia coli. J. Mol. Biol.





~r E and Cpx control synthesis of protein-folding enzymes

231: 261-273. Sambrook, J., E.F. Fritsch, and T. Maniatis. 1989. Molecular

cloning: A laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

Silhavy, T.J., M.L. Berman, and L.W. Enquist. 1984. Experi- ments with gene fusions. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

Simons, R.W., F. Houman, and N. Kleckner. 1987. Improved single and multicopy lac-based cloning vectors for protein and operon fusion. Gene 53: 85-96.

Slauch, J.M. and T.J. Silhavy. 1991. cis-acting ompF mutations that result in OmpR-dependent constitutive expression. J. Bacteriol. 173: 4039-4048.

Snyder, W.B. and T.J. Silhavy. 1995. f~-Galactosidase is inacti- vated by intermolecular disulfide bonds and is toxic when secreted to the periplasm of Escherichia coil [. Bacteriol. 177: 953-963.

Snyder, W.B., L.J.B. Davis, P.N. Danese, C.L. Cosma, and T.J. Silhavy. 1995. Overproduction of NlpE, a new outer-membrane lipoprotein, suppresses the toxicity of periplasmic LacZ by activation of the Cpx signal transduction pathway. J. Bacteriol. 177: 4216-4223.

Strauch, K. and J. Beckwith. 1988. An Escherichia coli mutation preventing degradation of abnormal periplasmic proteins. Proc. Natl. Acad. Sci. 85: 1576-1580.

Wang, Q.P. and J.M. Kaguni. 1989. A novel sigma-factor is involved in expression of the rpoH gene of Escherichia coli. J. BacterioI. 171: 4248-4253.





10.1101/gad.11.9.1183Access the most recent version at doi: 11:1997, Genes Dev.

P N Danese and T J Silhavy synthesis of periplasmic protein-folding enzymes in Escherichia coli.The sigma(E) and the Cpx signal transduction systems control the

References

http://genesdev.cshlp.org/content/11/9/1183.full.html#ref-list-1

This article cites 30 articles, 15 of which can be accessed free at:

License

ServiceEmail Alerting

click here.right corner of the article or

Receive free email alerts when new articles cite this article - sign up in the box at the top

Copyright © Cold Spring Harbor Laboratory Press


http://genesdev.cshlp.org/lookup/doi/10.1101/gad.11.9.1183

http://genesdev.cshlp.org/content/11/9/1183.full.html#ref-list-1

http://genesdev.cshlp.org/cgi/alerts/ctalert?alertType=citedby&addAlert=cited_by&saveAlert=no&cited_by_criteria_resid=protocols;10.1101/gad.11.9.1183&return_type=article&return_url=http://genesdev.cshlp.org/content/10.1101/gad.11.9.1183.full.pdf



the e and the cpx signal transduction systems control the...

Documents